ITT TCA350X BBD delay datasheet or pinout

Hi, I purchased a couple of vintage TCA350X analog delay lines using BBD, these are dip14 chips. Im having trouble locating the datasheet. The only thing I could find is the datasheet for the TCA350Y which is a DIP8 part http://www.introni.it/pdf/ITT%20-%20Integrated%20Circuits%20for%20Consumer%20Applications%201977_1978.pdf on page 170. Im guessing that the TCA350X is a dual version of the TCA350Y, but I need to know the pinout. In that same document theres a TCA380, which is a similar part and has 12pins.

Does anyone has the datasheet for the 350X? at least I would like to know the pin-out, that should be enough.

Thanks

user 37518 said:

Hi, I purchased a couple of vintage TCA350X analog delay lines using BBD, these are dip14 chips. Im having trouble locating the datasheet. The only thing I could find is the datasheet for the TCA350Y which is a DIP8 part http://www.introni.it/pdf/ITT%20-%20Integrated%20Circuits%20for%20Consumer%20Applications%201977_1978.pdf on page 170. Im guessing that the TCA350X is a dual version of the TCA350Y, but I need to know the pinout. In that same document theres a TCA380, which is a similar part and has 12pins.

Does anyone has the datasheet for the 350X? at least I would like to know the pin-out, that should be enough.

Thanks

Click to expand...

The pinout is listed in the data sheet within that link you provided.

1-leave vacant
2-clock input t2
3-delay line input
4-ground, 0
5-clock input t1
6-delay line output
7-Vdd
8-NC

I thought I was pretty familiar all the different flavors of BBD/CCD in the market but that is a new one for me, and I started messing with BBD delay back in the 70s.

From first glance I am not impressed. It lacks the redundant output stage to reduce the significant clock frequency energy of only a single output (signal is only present for 1/2 the clock period). So output filtering is much more difficult with all that clock energy injected into the audio path. It is only 185 stages long so not very long delay and wants -22v supply.  Even the experimental Phillips BBDs from the early 70s used the dual output structure, that everybody else used also (for obvious benefit).

JR

JohnRoberts said:

The pinout is listed in the data sheet within that link you provided.

1-leave vacant
2-clock input t2
3-delay line input
4-ground, 0
5-clock input t1
6-delay line output
7-Vdd
8-NC

I thought I was pretty familiar all the different flavors of BBD/CCD in the market but that is a new one for me, and I started messing with BBD delay back in the 70s.

From first glance I am not impressed. It lacks the redundant output stage to reduce the significant clock frequency energy of only a single output (signal is only present for 1/2 the clock period). So output filtering is much more difficult with all that clock energy injected into the audio path. It is only 185 stages long so not very long delay and wants -22v supply.  Even the experimental Phillips BBDs from the early 70s used the dual output structure, that everybody else used also (for obvious benefit).

JR

Click to expand...

Dear John, the pinout you provided is for the DIP8 version (the TCA350Y), I need pinout for the DIP14 version (TCA350X), I need that delay specially because it has short delay times, I need something in the usec range, whilst most BBD's are in the ms range.

From post 7 here
https://www.diystompboxes.com/smfforum/index.php?topic=84735.0
it looks it is a single BBD also, pinout is there.

I found this with google on some random forum so can't attest to accuracy

pin 2 : t2
pin 3 : IN
pin 4 : V(+)
pin 5 : t1
pin 6 : Out
pin 7 : V(-)

14PIN

pin 3 : t2
pin 4 : IN
pin 5 : V(+)
pin 10 : t1
pin 11 : Out
pin 12 : V(-)

Good luck...  500kHz looks like upper clock limit but we ran BBDs up to MHz clocks, but they get a little squirrely.

JR

Thank you both, I'll try that combination. Still missing a few pins but thats better than nothing.

If you need TCA350X pinout, you can find it on page 2 in this romanian article:

http://www.qsl.net/yo4tnv/docs/Tehnium/9706c.pdf

The datasheet can be downloaded here, but I think it's not complete:

https://datasheet4u.com/datasheet-pdf/ETC/TCA350/pdf.php?id=942038

185 stages is short, even for Karplus Strong type applications.
So ITT even made their own BBD, rather than just rubberstamp their own P/N on there? I have a few TDA stamped BBD's, but those are "just" rebranded Panasonics.

Jarno said:

185 stages is short, even for Karplus Strong type applications.
So ITT even made their own BBD, rather than just rubberstamp their own P/N on there? I have a few TDA stamped BBD's, but those are "just" rebranded Panasonics.

Click to expand...

The Panasonic (Matshusita) BBDs were licensed from Phillips design and used the dual output stage to reduce clock frequency noise.

JR

Jarno said:

185 stages is short, even for Karplus Strong type applications.
So ITT even made their own BBD, rather than just rubberstamp their own P/N on there? I have a few TDA stamped BBD's, but those are "just" rebranded Panasonics.

Click to expand...

I happen to need short stages for small delays, I need delays in the usec range, even 185 stages is a bit too much, I need it for analog computing in case you were wondering why I need such short delays.

On the other hand, the pinout that several of you posted seems to be correct, the thing that bothered me is that so many pins of the 14 pin package are unused, and I thought that the dip14 version (TDA350X) should include 2 delay lines inside because there is a dip8 (TDA350Y) version which includes only one delay, but no, it seems theres only one dela line in the 14 pin package and all the remaining 8 pins are unused.

Some of the BBD packages have those unused legs cut off, see MN3005.
Maybe roll your own BBD discreetly? Then you can choose supply voltage yourself, careful layout will be needed to reach high clock speeds though.

JohnRoberts said:

The Panasonic (Matshusita) BBDs were licensed from Phillips design and used the dual output stage to reduce clock frequency noise.

JR

Click to expand...

So the other way around, interesting.

Jarno said:

So the other way around, interesting.

Click to expand...

Yup, in the early 1970s the company I worked for was using experimental early BBDs from Phillips to generate a pitch shifted audio***, to you could speed up playback of talking books on tape, and restore the pitch back down to normal.  We called this "time compression" for shifting pitch down, "time expansion" for shifting pitch up. As I recall Matshushita licensed the technology and sold them commercially under Panasonic brand.

Back in 1976 I sold a delay line kit in Popular Electronics that used a Panasonic BBD. Later US company Reticon made a BBD version using opposite polarity MOS technology from Panasonic.

JR

*** By clocking a BBD delay line with ramping clock frequency (saw tooth driving V to period  convertor) will stretch out, or squeeze together the waveform passing through the delay line, effectively shifting the pitch up or down. Once per delay cycle when you jump back from low to high (or high to low) clock frequency you must discard (blank out) the first 256 invalid samples.  High technology for early 70s and cheaper then digital processing of the day that was struggling to realize 8 bit resolution. 

If you clock the delay line with a triangle wave instead of a saw tooth, you get an alternating pitch step up and step down. Using two delay lines with alternate up/down cycles you can generate 3 pitches (normal + one up and one down). I've used this and seen others in studio efx. 

JohnRoberts said:

*** By clocking a BBD delay line with ramping clock frequency (saw tooth driving V to period  convertor) will stretch out, or squeeze together the waveform passing through the delay line, effectively shifting the pitch up or down. Once per delay cycle when you jump back from low to high (or high to low) clock frequency you must discard (blank out) the first 256 invalid samples.  High technology for early 70s and cheaper then digital processing of the day that was struggling to realize 8 bit resolution. 

If you clock the delay line with a triangle wave instead of a saw tooth, you get an alternating pitch step up and step down. Using two delay lines with alternate up/down cycles you can generate 3 pitches (normal + one up and one down). I've used this and seen others in studio efx.

Click to expand...

That sounds really interesting, I should definitely play around with these.

Jarno said:

Maybe roll your own BBD discreetly? Then you can choose supply voltage yourself, careful layout will be needed to reach high clock speeds though.

Click to expand...

I have thought of rolling my own BBD for shorter time delays using a lower frequency, Nelson Pass gave me the inspiration with his "Thousand FET beast" amplifier.

I have sent you a pm, user 37518!
Cheers
Bernd

user 37518 said:

I happen to need short stages for small delays, I need delays in the usec range, even 185 stages is a bit too much, I need it for analog computing in case you were wondering why I need such short delays.

Click to expand...

What kind of performance do you need? It may be that using cascaded all-pass filters would be more suitable.
One cell can produce 200ns of delay up to 100kHz.

abbey road d enfer said:

What kind of performance do you need? It may be that using cascaded all-pass filters would be more suitable.
One cell can produce 200ns of delay up to 100kHz.

Click to expand...

Yes, I opted for the All-pass sections.